search

uchicago-computation-workshop / tor_wager

Repository for Tor Wager's presentation at the CSS Workshop (2/21/2019)
1 stars 0 forks source link

Choice of decision rule in testing (and other questions) #20

Open jsgenan opened 5 years ago

jsgenan commented 5 years ago

Thank you for bringing an interesting paper! Bayesian model fits the intuition perfectly, but I might need some backgrounds for understanding the model better.

  1. The inter-stimulus interval is different in two studies. What's the consideration behind this design? My concern is that the interval might intervene how strong the perception lasts.
  2. I noticed that when graphing the brain activity, you are using both FDR and bootstrap p value. What's the implication of using two decision rules? Since here N is small, how does FDR q<0.05 compare with p<0.01? Thanks again!